But different prices lead to different levels of production. And there is in fact an optimal level of production of a public good. Certainly in a public bid system where people overbid they are purchasing two different goods, pride (or spite, or advertising for their business, or whatever) and the public good. But this is also true when people underbid. They are purchasing the public good and the goods they can purchase with the money below the optimal amount. There is one price that the assurance firm will charge that will lead to the optimal amount of good produced, and anything above or below will lead to too much or too little of the public good being produced.

This doesn't have to be true.

Pretend the world only has money and this one good. Avoid talking about the value of pride, opportunity costs, etc. Until the simple situation is resolved, adding in these (irrelevant) extra considerations just makes it harder to discuss the fundamental issue.

How so?

That sounds pretty boring to me. I'm not sure what there would be to discuss.

See my earlier post:

Price doesn't matter. Levels of production do. If I buy something from you, we are both better off. What price we settle on impacts how much better off we are individually (a higher price means that you capture more of the surplus).

If you value the object less than I do, and I buy if from you, we reach the socially optimal result, regardless of price (in particular, even if I pay more than what the good is worth to me). Otherwise (i.e., no transaction), the outcome is inefficient.

At any rate, we aren't talking about public goods. Time for a "Markets, Efficiency, and Rationality: Containment Thread".

Well, to be blunt, I don't think you really understand the situation with money and one good very well. As a result you make errors in the more complicated situation that you would also make if you were analyzing the simper situation. But it's easier for you to detect (and for me to explain) those errors in the toy case.

Why do you think I don't understand the toy case?

To start:

1. Discussion of price and efficiency
2. Treatment of private values
3. Trying to avoid fundamental problems by adding more goods

I'm also not sure you have a deep understanding of why the "free rider problem" (e.g., underproduction) exists without externality, or what information economists capture with price (i.e., all of it).

I don't want to come across as arrogant here, or even obnoxious. The toy case is already very difficult and has many subtleties. That's why I'd prefer to discuss it rather than more complicated situations where the fundamental questions are largely the same but there are more distractions.

He thinks this because you keep making definitional mistakes and introducing irrelevant additional variables.

Of course I understand how the free rider problem can exist without externality. If someone values the good at a dollar but pays a cent he is not benefiting from an externality because he is a party to the transaction.

What fundamental problems do you think I am attempting to avoid by adding more goods than money and the public good?

Since I have made definitional mistakes why don't you help me out by pointing them out and correcting them.

What do you think I've been trying to do, dude? Sholar and ShaneP are engaged in the same. If you follow the trail in our posts collectively, the answers are there.

I have followed all the posts in this thread. ShaneP seems to have agreed with the point I was trying to make about how people overdonating in order to purchase both the public good and some other good like advertising for their business or pride or spite would lead to overproduction of a public good. If you think I'm making a fundamental mistake please elaborate on what it might be.

What good is overproduced? How do you know it is overproduced? (Be specific with your example.)

Now imagine that instead of "pride", they got a candy bar.
What good is overproduced? How do you know it is overproduced?

If anything about your answer changes, you are making a mistake.

OK let me say that I am assuming that the public good is produced in direct proportion to the revenue the assurance firm generates. I am also assuming that there exists a specific amount of the public good that is the optimal amount. It is not measurable what that amount is but some amount is socially optimal. Since the good is produced in direct proportion to the revenue of the assurance firm, there is only one value of revenue that will produce the optimal amount of good.

So imagine two scenarios. In both scenarios the assurance firm must get everyone who would use the public good to agree to contribute. In the first the bids are secret, and in the second the bids are published. I think it is fairly uncontroversial to say that the revenues would be expected to be higher in the second scenario, thus more of the public good would be produced. This is because in the second scenario, two things are being sold: pride (a private good) and the public good itself.

Now let's say that a utilitometer is introduced to analyze the situation and it is found that in the secret bid scenario the public good is underproduced. Hardly surprising. In the public bid situation it is found that the public good is overproduced. This may or may not be a realistic situation, depending on the value people would put on the good "pride."

So you say might say, sure, in the public bid situation the public good was overproduced, but it is actually an efficient tranaction because people got X dollars worth of pride in surplus of the value of the public good.

I think this is faulty reasoning however. Let's take citizen Tom as an example. Let's say that the utilitometer shows that Tom values the public good at 100 dollars. In the private bid situation he bids 50 dollars. In the public bid situation he bids 150 dollars. We can go ahead and say that Tom is clairvoyant and accurately knows how much he values the public good. In the private bid situation because Tom underbids, the public good is underproduced by 50 dollars. While Tom gets to keep his 50 dollars, we know he would value the public good more. In the overbidding case, Tom pays 150 dollars and the public good is overproduced. But Tom gets 50 dollars worth of pride. But he pays for that pride in 50 dollars worth of excessive public good.

To me the overbidding and underbidding are equivalent. Tell me why I'm wrong.

Simply put, the underbidding is because of the characteristics of the good. The overbidding has nothing to do with the good being provided, and is a result of the mechanism that you're using to solicit bids.

Note: I'm not exactly sure what you mean by 'equivalent' though. It seems like it's shorthand for ... something. Equivalent in what sense?

One is exactly as harmful to society as another. Assuming a symmetric utility function around the socially optimal amount of good produced I suppose.

Ok, I think I agree with this as well. But to play devil's advocate (and I also agree with the thrust of this argument), there are very high transactions costs to this system. If the assurance firm is able to accurately do research on the cost/value of a given task, but the transaction costs of convincing people of said value are too high, then we actually end up with massive underproduction. Whereas if there was a collective pool of money and an appointed assurance firm to simply make decisions...

This is already a weird and unnecessary assumption. It causes most of what follows to fall apart. Consider the case where the good in question is not continuous. In fact, consider the case where it is binary: some good is either produced or not produced, there is no "quantity" to speak of.

Note that this depends on the cost of providing the good.

This is where the earlier assumption starts to cause problems. Recall that questions of pricing and how the surplus gets split are generally irrelevant here.

Why? This is definitely a false or unnecessary assumption.

Again, this is not correct, though it follows from the erroneous assumption. Rethink the situation where the good is either produced or not produced.

You don't need a utilitometer to determine whether or not the good is underproduced. This is extremely important to understand.

In fact I'm no longer sure that we are using the word "underproduced" in the same way.

Almost as if they got a candy bar instead of pride...

Why does he do this?

This has nothing to do with clairvoyance. Tom, as an economic agent, can price bundles of goods. (The price where he is indifferent between the money and the goods.) He need only know his own private values in order to do this.

This doesn't make sense.

He paid for the candy bar pride with dollars.

See the comments above.

In the case where Tom underbids by 50 dollars he can use that 50 dollars to buy candy bars.

Answers like this should explain why I prefer to start with the single-good situation.

Alternatively: The reason I asked you to consider "candy bars" instead of "pride" is so that you would realize the above. Your analysis of the situation did not take this possibility into account.

Also,

I know that not all public goods would be produced directly in proportion to the revenue of the assurance firm, and that some in fact could be binary. This is why I say I am assuming that the good is produced in proportion to revenue. Funny that you should criticize me for introducing extra variables and then when I try to create a toy example with certain assumptions you say "those assumptions aren't necessarily true!" Well, no ****, that's why I said assume they are true.

Of course I realized this.

Of three scenarios, the two I describe in my example and a third scenario where the exactly optimal amount of public good is created, am I correct in saying the the total utility of the public is highest in the third scenario and equal in the first two? Assuming that the monetary value of the overproduction and underproduction is the same.

Heh. Perhaps government is necessary, but one thing to consider is that having a government produce public goods can have a stifling effect on innovations in private sector public goods production which would tend to lower those transaction costs. So us abstractly talking about how assurance contracts might work and trying to estimate how large the transaction costs might be is sort of like people in 1980 trying to talk about the problems involved in iPad production.

Fair enough, but that assumption is still a poor one in that it gives your example a lot of features which are not in the general case

The importance of starting with the right toy example (i.e., what can be assumed, what can be ignored, and what to focus on) cannot be overstated.

Rather than to attempt to explain why my version is going to be more illuminating, I'll ask for you to trust me and think about the binary situation for the time being.

"Total utility" doesn't mean anything.

Nothing is overproduced in your example (Tom just buys his candy bars at the same time as his public good, for example).

Isn't maximization of utility how the optimal production of a public good is defined?

But I am correct in saying that in the secret bid scenario the public good is underproduced?

So Tom bought $50 worth of pride, which is equivalent to $50 of candy bars. Let's say everyone else in town did the same thing. Now the public good is overproduced by 50 dollars and everyone in town has 50 dollars of pride (same as candy bars) per person. Now as a group, wouldn't everyone have been better off if they had donated 100 dollars per person (let's assume everyone values the good the same as Tom), produced the optimal amount of public good and had the 50 dollars to spend on candy bars? Everyone is indifferent between 50 dollars worth of candy bars and pride, but they are not indifferent between the utility the 101st-150th dollars of production of the public good provides and the 50 dollars worth of candy bars.

But they didn't donate the optimal amount, because while the aggregate behavior of the group led to overproduction of the public good, any specific person donating in excess of 100 dollars will not change the utility realized from the public good. So they might lose a tiny amount by overpaying for the public good (in that their actions will be responsible for a tiny overproduction of the public good) but they will realize all 50 dollars worth of benefit from the pride they are buying.

Not at all. Maximization of the sum of consumer surplus and producer surplus is now the optimal level is defined. Utility enters absolutely nowhere.